Policy Pathways for Coordinated CO₂ and Air Pollutant Reductions in Urban Road Transport: A Case Study of Zhengzhou, China

Abstract

Urban road transport policies must simultaneously address climate mitigation, local air quality, and the infrastructure requirements associated with vehicle electrification. However, these dimensions are rarely evaluated within a unified city-level framework. This study develops an integrated assessment framework that combines a bottom-up co-source inventory of CO₂ and seven air pollutants, Long-range Energy Alternatives Planning (LEAP)-based scenario modeling, policy contribution analysis, elasticity-based co-benefit assessment, and electric vehicle charging demand estimation for Zhengzhou, China. In 2022, the road transport sector consumed 10,178 ktce of energy and emitted 27.8 Mt of CO₂. Private cars contributed 66.7% of CO₂ emissions, whereas heavy- and medium-duty trucks and light-duty trucks contributed 48.1% and 27.9% of NOx emissions, respectively, collectively accounting for 76.0% of the total. Under the existing policy scenario (EPS), CO₂ emissions increase to 45 Mt in 2030 and 55 Mt in 2040. Under the dual carbon scenario (DCS), emissions peak at approximately 36 Mt in 2030 and decline to 32 Mt by 2040, representing reductions of 20% and 42% relative to the EPS, respectively. Electric vehicle promotion and green transport development contribute 42% and 32% of peak-year CO₂ mitigation. Policy effectiveness differs across emission types. Electric vehicle promotion and green public transport are relatively more effective for CO₂ mitigation, whereas old vehicle retirement, motorcycle phase-out, light-truck electrification, and tighter emission standards provide greater air pollutant reduction benefits. Supporting an electric vehicle stock of approximately 1.22 million in 2030 would require about 610,000 charging piles at a vehicle-to-charger ratio of 2:1. The principal contribution of this study is to demonstrate how complementary vehicle technology, transport structure, emission control, power sector, and infrastructure policies can be combined to support city-level carbon peaking and air pollution co-control.

1. Introduction

Climate change and air pollution are two major global environmental challenges [1,2]. China has made substantial progress in the low-carbon energy transition and air pollution control. The country aims to realize the “Beautiful China” vision by 2035, attain peak carbon emissions prior to 2030, and reach carbon neutrality by 2060 while accelerating coordinated reductions in CO₂ and air pollutants under the dual goals of “carbon peaking” and “air quality improvement.” China’s transportation sector is a major energy consumer and emission source, consuming approximately 42.8% of national gasoline and 52.5% of diesel, while contributing about 9.1% to total energy-related CO₂ emissions. Rising vehicle ownership and freight volumes have significantly intensified both CO₂ and air pollutant emissions from road transport [3,4].

On-road vehicle emissions are commonly quantified using bottom-up, activity-based inventory methods [5,6,7]. Emissions are calculated by combining the vehicle population, annual vehicle kilometers traveled (VKT), and pollutant-specific emission factors differentiated by vehicle category, fuel type, emission standard, and operating condition. These inventories provide detailed temporal and spatial insights into mobile source pollution, aiding policymakers in developing targeted control measures. Source apportionment studies and emission inventories for Chinese cities indicate that road vehicles contribute 10–30% to urban PM_2.5 [8] and over 40% to NOx in megacities such as Beijing, Shanghai, and Guangzhou [9]. Private cars and motorcycles have been identified as major sources of urban air pollutant emissions [10]. However, most existing inventory studies have focused predominantly on criteria air pollutants, with relatively few efforts to develop integrated inventories that simultaneously account for CO₂ and air pollutant emissions from the same source categories.

Recent studies have increasingly examined the co-evolution of transport-related CO₂ and air pollutants and the mitigation potential of vehicle electrification, transport mode shifts, and tighter emission standards. Scenario-based projections for road transport emissions typically employ models such as the Long-range Energy Alternatives Planning (LEAP) system, combined with emission inventory data. Ou et al. (2010) utilized LEAP to project energy consumption and pollutant emissions from China’s road transportation system, highlighting increased oil demand due to vehicle growth and the benefits of alternative fuels [11]. Subsequent studies have similarly projected rising energy consumption and CO₂ emissions while evaluating the effectiveness of various control measures [12,13,14]. Liu et al. (2018) projected the energy consumption of China’s vehicle fleet from 2016 to 2050, identifying energy peaks by 2027 under strengthened scenarios involving new energy vehicles, traffic management, and green mobility [15]. Additional studies have demonstrated the potential for pollutant reductions through measures such as vehicle restrictions, stricter emission standards, accelerated retirement of high-emission vehicles, and green energy promotion [16,17,18,19]. Despite these advances, most studies have addressed CO₂ and air pollutant reductions in isolation, and integrated co-reduction frameworks remain scarce, particularly at the city level where policy implementation occurs.

An increasing body of research has focused on the environmental co-benefits of electric vehicles (EVs) given their rapid market penetration. Shang et al. (2024) assessed the impacts of large-scale EV promotion in Shanghai, demonstrating that NOx, PM, and CO₂ emissions could decrease by 4.94%, 22.51%, and 24.60%, respectively, in 2025 relative to 2018 levels [20]. Xu et al. (2023) reported cumulative greenhouse gas reductions of 6.62 million tons from 2020 to 2030 through vehicle electrification in Hainan Province [21]. Van Vliet et al. (2011) analyzed EV emissions across different operational cycles, finding that carbon reduction benefits improve and unit costs decline as the share of renewable electricity increases [22]. Collectively, these studies demonstrate that EV promotion can enable synergistic reductions in both carbon and air pollutant emissions. However, realizing this potential at scale requires robust charging infrastructure support.

Studies demonstrate that EV penetration correlates positively with the expansion of charging infrastructure [23,24]. Optimizing the integrated “vehicle–pile–grid–storage” system—through intelligent scheduling, efficient charger maintenance, grid demand response, and dynamic energy storage management—is essential for sustaining high-penetration EV charging efficiency [25,26]. Such optimization requires intelligent platforms capable of real-time data fusion, predictive scheduling, and distributed control [27]. China’s National Energy Administration has set a long-term target of a 1:1 vehicle-to-pile ratio, with a transitional 2:1 ratio expected post-2025 as the market matures [28].

As a rapidly motorizing inland metropolis and a major passenger and freight transport hub in central China, Zhengzhou combines a large private car fleet with intensive diesel freight activity and expanding electric mobility. This study therefore develops an integrated city-level framework to: (1) compile a co-source inventory of CO₂ and major air pollutants for 2022; (2) project energy consumption and emissions to 2040 under an existing policy scenario (EPS) and a dual carbon scenario (DCS) using the LEAP model; (3) quantify the contributions and co-benefits of individual mitigation measures; and (4) estimate the charging infrastructure required to support the projected electric vehicle fleet. By linking emissions, policy contributions, co-benefit indicators, and infrastructure demand within one analytical framework, this study provides a more operational basis for coordinated urban transport decarbonization and air pollution control.

2. Methods

2.1. Study Area

Zhengzhou, the capital of Henan Province, is located in central China (34°16′–34°58′ N, 112°42′–114°14′ E). In 2022, the city had a resident population of approximately 12.8 million and a gross domestic product of CNY 1.29 trillion. Zhengzhou is an important inland transport and logistics hub, with extensive expressway and railway connections, a major international airport, and national logistics and freight-train functions. Its large vehicle stock, rapidly growing passenger travel demand, and intensive intercity freight activity make it a representative case for assessing coordinated CO₂ and air pollutant mitigation in rapidly motorizing inland cities. The analysis integrates data from the Zhengzhou Statistical Yearbook, vehicle registration records, national and local technical guidelines, previous Zhengzhou vehicle emission studies, fuel and electricity parameters, and national, provincial, and municipal policy documents. The geographical location of Zhengzhou is shown in Figure 1.

2.2. Methodological Framework

The methodological framework consists of four linked modules. First, a bottom-up co-source emission inventory was developed for Zhengzhou’s road transport sector in the 2022 base year. The inventory covers CO₂ and seven air pollutants, including CO, NOx, volatile organic compounds (VOCs), PM₁₀, PM_2.5, SO₂, and NH₃, across seven vehicle categories and multiple emission standards. Second, the base-year inventory was incorporated into the LEAP model to project vehicle stock, energy demand, CO₂ emissions, and exhaust air pollutant emissions from 2022 to 2040 under the EPS and DCS. Third, the contributions of individual mitigation measures were quantified for the carbon-peaking year, and elasticity coefficients were used to characterize the relative co-benefits between CO₂ and air pollutant reductions. Fourth, future charging infrastructure demand was estimated from the projected electric vehicle stock and the target vehicle-to-charger ratio.

The accounting boundary differs between CO₂ and local air pollutants. For gasoline, diesel, and liquefied-petroleum-gas vehicles, operational CO₂ emissions are calculated from fuel consumption, fuel carbon content, and carbon oxidation. For EVs, tailpipe CO₂ emissions are zero, while indirect CO₂ emissions from electricity generation are calculated by multiplying the vehicle electricity consumption by the corresponding grid electricity emission factor. The projected decarbonization of the electricity system is incorporated through scenario-specific electricity emission factors. Local air pollutant emissions include vehicle exhaust emissions within the road transport sector. Upstream air pollutant emissions from electricity generation and fuel production, vehicle and battery manufacturing, infrastructure construction, and non-exhaust particulate emissions are outside the present system boundary and are discussed as limitations.

2.3. Inventory of Air Pollutants and CO₂ Emissions

The 2022 road transport emission inventory was established using a bottom-up activity-based approach. Emissions were calculated by multiplying the vehicle population, annual vehicle kilometers traveled, and pollutant-specific emission factors for each vehicle category and emission standard. The vehicle categories included heavy- and medium-duty trucks, light-duty trucks, buses, taxis, private cars, motorcycles, and other vehicles. The emission standards covered Pre-China I to China VI vehicles in use in 2022.

2.3.1. Vehicle Emissions

For each pollutant, the annual emissions from vehicle category i and emission standard j were calculated as follows:

$${\mathrm{E}}_{\mathrm{p}}={\displaystyle {\sum }_{\mathrm{j}}^{\mathrm{i}}}{\mathrm{V}\mathrm{P}}_{\mathrm{i},\mathrm{j}}\times {\mathrm{V}\mathrm{K}\mathrm{T}}_{\mathrm{i}}\times {\mathrm{E}\mathrm{F}}_{\mathrm{i},\mathrm{j},\mathrm{p}}\times {10}^{-6}$$

(1)

where ${\mathrm{V}\mathrm{P}}_{\mathrm{i},\mathrm{j}}$ denotes the vehicle population of type i under emission standard j; ${\mathrm{V}\mathrm{K}\mathrm{T}}_{\mathrm{i}}$ represents the annual average distance traveled (km); and ${\mathrm{E}\mathrm{F}}_{\mathrm{i},\mathrm{j},\mathrm{p}}$ represents the pollutants and CO₂ emission factor (g/km). The pollutants considered include CO, NOx, VOCs, PM₁₀, PM_2.5, SO₂, NH₃, and CO₂. The emission standards for vehicles include Pre-China I to China VI standards in use in 2022.

2.3.2. Vehicle Population by Emission Standard

Vehicle populations disaggregated by emission standard are not directly available from statistical data. This study estimates them based on registration timelines and survival rates, assuming that newly registered vehicles comply with the prevailing emission standard at the time of registration [29]. Historical and annual new registration data for Zhengzhou were obtained from the Zhengzhou Statistical Yearbook. The vehicle population by emission standard was calculated using Equation (2):

$${\mathrm{P}}_{\mathrm{i}} ={\sum }_{\mathrm{j} = 0}^{6 }\sum {\mathrm{N}}_{\mathrm{i}, \mathrm{m}, \mathrm{n}},$$

(2)

where j represents the applicable vehicle emission standard; n is the number of years during which emission standard j has been in effect for vehicle type I; m is the year in which emission standard j was first introduced; P denotes the total vehicle stock; and N is the adjusted count of newly registered vehicles accounting for survival rates.

Modified vehicle stock N was estimated using the correction formula (3):

$${\mathrm{N}}_{\mathrm{h},\mathrm{k}} = {\mathrm{R}}_{\mathrm{h}} \times { \mathrm{S}}_{\mathrm{h},\mathrm{k}-\mathrm{h}} ,$$

(3)

Here, ${\mathrm{N}}_{\mathrm{h},\mathrm{k}}$ represents the portion of vehicles registered in year $\mathrm{h}$ that are still in service in the target year k after applying the survival adjustment; ${\mathrm{R}}_{\mathrm{h}}$ denotes the number of new registrations in year $\mathrm{h}$ (with 2022 taken as the reference year); and ${ \mathrm{S}}_{\mathrm{h},\mathrm{k}-\mathrm{h}}$ corresponds to the survival percentage in the target year for vehicles that have been in use for k − h years.

Survival rates followed Zachariadis et al. (1995) [30] using Equation (4):

$${\mathrm{S}}_{\mathrm{h}, \mathrm{k}-\mathrm{h}} = \mathrm{EXP}[-{(\frac{\mathrm{k} + \mathrm{c}}{\mathrm{T}})}^{\mathrm{c}}],$$

(4)

In this formulation, T and c characterize the expected operational lifetime of a vehicle. For passenger vehicles, T = 32 and c = 7. For trucks, T = 25 and c = 10.

2.3.3. Vehicle Kilometers Traveled

The VKT directly influences emission calculations. In this study, the VKT values for each vehicle category were primarily sourced from the “Technical Guide for Compiling Air Pollutant Emission Inventories from On-Road Motor Vehicles” and calibrated against local data from a prior Zhengzhou study [31], as presented in Table S1. Seven vehicle categories are considered: heavy- and medium-duty trucks (H&M trucks), light-duty trucks, buses, taxis, private cars, motorcycles, and others.

2.3.4. Localized Emission Factors for Air Pollutants and CO₂

Local emission factors for criteria air pollutants were derived by adjusting national baseline factors to reflect Zhengzhou-specific conditions, including meteorological parameters, vehicle fleet composition, fuel types, and emission standards (Pre-China I through China VI). The localized emission factor for each pollutant was calculated using Equation (5):

$${\mathrm{EF}}_{ \mathrm{p}, \mathrm{i}, \mathrm{j}}={ \mathrm{BEF}}_{\mathrm{i}, \mathrm{j}, \mathrm{p}} \times \mathrm{\phi } \times { \mathrm{\gamma }}_{\mathrm{p}, \mathrm{i}, \mathrm{j}} \times {\mathrm{\mu }}_{\mathrm{p}, \mathrm{i}, \mathrm{j}} \times {\mathrm{\theta }}_{\mathrm{i}}$$

(5)

where ${\mathrm{EF}}_{ \mathrm{p}, \mathrm{i}, \mathrm{j}}$ is the localized emission factor for pollutant p, emission standard j, vehicle type i in Zhengzhou; BEF is the baseline emission coefficient; φ is the meteorological correction factor (temperature and humidity, as presented in Table S2); γ and μ are the average speed and deterioration correction coefficients, respectively; and θ represents other emission correction factors.

$${\mathrm{EF}}_{\mathrm{f}}=\mathrm{NCV} \times {\mathrm{FCR}}_{\mathrm{f}} \times \mathrm{\rho } \times \mathrm{CC} \times \mathrm{COF} \times {\displaystyle \frac{44}{12}}$$

(6)

where ${ \mathrm{EF}}_{\mathrm{f}}$ denotes the CO₂ emissions per 100 km for the specified vehicle. f represents the fuel type, NCV is the lower heating value of the fuel (GJ/t), FCR represents the fuel consumption (L/100 km), ρ corresponds to the fuel density (g/L), CC expresses the carbon content (tC/GJ), COF is the carbon oxidation rate (99%), and 44/12 is the molecular weight ratio of CO₂ to elemental carbon.

2.4. LEAP Model Structure

The LEAP model was configured to represent Zhengzhou’s road transport sector using a hierarchical activity-based structure. The transport sector was divided into passenger transport and freight transport. Passenger transport included private cars, buses, taxis, motorcycles, and other vehicles, while freight transport included heavy- and medium-duty trucks and light-duty trucks. Each vehicle category was further classified by fuel type, including gasoline, diesel, liquefied petroleum gas, and electricity, and by emission standard from Pre-China I to China VI. The 2022 emission inventory was used as the base-year calibration dataset. The key model inputs included the vehicle stock, annual VKT, fuel consumption intensity, energy type, emission-standard distribution, localized emission factors, electricity emission factors, and policy-driven changes in vehicle structure and travel activity. The model outputs included annual energy consumption, CO₂ emissions, and air pollutant emissions from 2022 to 2040 under each scenario. Model consistency was checked by comparing the calculated 2022 energy consumption and emissions with local statistical data and previous Zhengzhou emission studies where available.

2.5. Scenario Design

Two scenarios were developed for 2022–2040. The EPS represents the continuation of enacted transport, vehicle emission, energy efficiency, and electrification policies. The dual carbon scenario (DCS) represents an enhanced mitigation pathway consistent with China’s carbon-peaking and carbon-neutrality objectives. Both scenarios use the same calibrated 2022 vehicle stock, activity, energy, and emission data but differ in vehicle electrification, fuel-economy improvement, retirement of high-emitting vehicles, transport activity, freight modal shift, and electricity emission intensity. Unless otherwise specified, annual values between 2022, 2030, and 2040 were obtained through linear interpolation.

The scenario assumptions were developed primarily from Chinese national, provincial, and local policy directions, including the New Energy Vehicle Industry Development Plan (2021–2035), the Action Plan for Carbon Dioxide Peaking Before 2030, Henan Province’s carbon-peaking and transport-development policies, and local measures for public transport, vehicle retirement, logistics electrification, and freight modal shifts. International experience is used only as supplementary context. The DCS assumptions represent an ambitious analytical pathway rather than formally adopted Zhengzhou targets.

The provincial average electricity CO₂ emission factor for Henan in 2022 was 0.6058 kg CO₂/kWh. To represent the gradual decarbonization of the power sector, the emission factor was assumed to decline annually by 1% under the EPS and 1.5% under the DCS. The resulting electricity emission factors were 0.5590 and 0.5055 kg CO₂/kWh in 2030 and 2040 under the EPS, and 0.5368 and 0.4615 kg CO₂/kWh under the DCS, respectively. The 1% and 1.5% annual decline rates represent moderate and enhanced power sector decarbonization assumptions under the EPS and DCS, respectively.

Energy structure adjustment. According to China’s Action Plan for Carbon Peak Before 2030, the transport sector should accelerate vehicle electrification and reduce dependence on conventional fuels. Zhengzhou registered approximately 116,000, 124,000, and 140,000 new vehicles from 2022 to 2024, of which 31%, 26%, and 25% were new energy vehicles (NEVs), respectively. Based on these trends, NEVs are assumed to account for 40% and 70% of new vehicle registrations in 2030 and 2040 under the EPS, and 60% and 100% under the DCS. The DCS represents an enhanced electrification pathway consistent with national decarbonization goals and international experience.

Energy efficiency improvement. Following national air pollution control requirements, diesel vehicles below the China III standard and gasoline vehicles below the China I standard are progressively retired. Both scenarios apply these phase-out requirements, while the DCS assumes faster implementation and replacement with vehicles meeting stricter emission standards.

Transport structure optimization. National carbon-peaking and pollution-control policies encourage public transport development, intelligent mobility, multimodal transport, and the transfer of medium- and long-distance freight from road to rail and waterway transport. Under the EPS, road freight and passenger transport activity are assumed to decrease by 25% and 40%, respectively, by 2040. Under the DCS, total road freight and passenger vehicle kilometers traveled are assumed to decline by 40% and 60%, respectively, relative to 2022. In addition, all 6316 buses in Zhengzhou are assumed to be electrified from 2022 under the DCS, in accordance with the provincial implementation plan.

2.6. Co-Benefit Assessment Using Elasticity Coefficients

Elasticity coefficients (ELSs) are employed in this study to evaluate whether a given mitigation measure generates synergistic co-reductions in CO₂ and criteria air pollutants and to quantify the relative magnitude of such co-benefits. The ELS indicates how sensitive the proportional change in CO₂ emissions is to the proportional change in a specific pollutant when a particular control measure is implemented. It is defined as:

$${\mathrm{E}\mathrm{L}\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2/\mathrm{n}}={\displaystyle \frac{∆{\mathrm{\phi }}_{{\mathrm{C}\mathrm{O}}_2}/{\mathrm{\phi }}_{{\mathrm{C}\mathrm{O}}_2}}{∆{\mathrm{\phi }}_{\mathrm{n}}/{\mathrm{\phi }}_{\mathrm{n}}}}$$

(7)

where ${\mathrm{\phi }}_{{\mathrm{C}\mathrm{O}}_2}$ and φ_n denote the total emissions of CO₂ and pollutant $\mathrm{n}$, respectively, and $∆{\mathrm{\phi }}_{{\mathrm{C}\mathrm{O}}_2}$ and ∆φ_n represent the emission reductions achieved through a specific policy measure. The ELS values are interpreted as follows:

• ELS = 0: The measure reduces only the pollutant with no effect on CO₂ emissions, indicating no synergistic effect;
• ELS < 0: The measure reduces one type of emission while increasing the other (a trade-off relationship), indicating a negative synergistic effect;
• 0 < ELS < 1: The measure achieves synergistic reduction, with a proportionally greater effect on the pollutant than on CO₂;
• ELS = 1: CO₂ and pollutant emissions decline at the same proportional rate;
• ELS > 1: The measure achieves synergistic reduction, with a proportionally greater effect on CO₂ than on the pollutant.

When the pollutant reduction is zero while CO₂ emissions decrease, the ELS is mathematically undefined and is reported as NA. When the CO₂ reduction is zero while the pollutant decreases, ELS equals zero.

2.7. Charging Infrastructure Demand Assessment

The penetration rate of EVs is a calculation parameter based on the total amount of vehicles, which accurately reflects the actual popularity of electric vehicles within the entire transportation system and their long-term cumulative impact on the energy structure, grid load, and carbon emissions. The rate is a stock concept, with relatively slow changes, and better reflects the macro process and long-term achievements of energy transformation. It is the basis for evaluating the demand for charging infrastructure, grid renovation planning, and long-term environmental benefits [32]. Its calculation formula is:

$$\mathrm{\mu }={\displaystyle \frac{\mathrm{E}\mathrm{V}\mathrm{s}}{\mathrm{V}\mathrm{p}}}$$

(8)

The vehicle-to-charger ratio (VCR) is a widely used indicator for evaluating the supply–demand balance of charging infrastructure:

$$\mathrm{V}\mathrm{C}\mathrm{R}={\displaystyle \frac{\mathrm{E}\mathrm{V}\mathrm{s}}{\mathrm{V}\mathrm{C}\mathrm{s}}}$$

(9)

where EVs is the number of electric vehicles and VCs is the number of installed charging piles. Based on a target VCR (VCRp), the future charging infrastructure demand (VCD) can be estimated as:

$$\mathrm{V}\mathrm{C}\mathrm{D}={\displaystyle \frac{\mathrm{E}\mathrm{V}\mathrm{s}}{{\mathrm{V}\mathrm{C}\mathrm{R}}_{\mathrm{p}}}}$$

(10)

This study adopts a target VCRp of 2:1, consistent with the post-2025 planning benchmarks set by China’s National Energy Administration and comparable to the targets adopted in Chongqing and Shenyang [33,34].

3. Results and Discussion

Zhengzhou’s road transportation emissions were characterized for the base year 2022, including vehicle stock, energy consumption, and emissions. Using the LEAP model, emission trends for CO₂ and conventional pollutants were projected from 2022 to 2040 under existing policy and dual carbon scenarios. Additionally, charging infrastructure demand was assessed for high EV penetration.

3.1. Characteristics of Vehicle Population and Emissions in 2022

Figure 2 summarizes the vehicle fleet composition, energy consumption structure, and associated CO₂ emissions of Zhengzhou’s road transport sector in 2022. The total energy consumption was 10,178 ktce, with gasoline being the dominant fuel at 62.3%, followed by diesel at 31.1%, while electricity accounted for only 0.6% (Figure 2a). Private cars were the largest energy consumers, accounting for 55.2% of total energy consumption, followed by the freight sector at 35.4%, with H&M trucks and light-duty trucks contributing 21.7% and 13.7%, respectively (Figure 2b). Total CO₂ emissions amounted to 27,771 kt, of which private cars contributed 66.7%, followed by H&M trucks (16.3%) and light-duty trucks (9.0%) (Figure 2c). The contrasting source profiles of CO₂ and NOx reflect fundamental differences between passenger and freight transport. Private cars dominate CO₂ emissions because of their large fleet size and aggregate travel demand. Heavy- and medium-duty trucks and light-duty trucks contribute disproportionately to NOx because of their diesel dependence, higher engine loads, and higher distance-specific emission factors. A uniform vehicle-control strategy would therefore be inefficient. Passenger-car policies should prioritize travel-demand management, electrification, and electricity decarbonization, whereas freight policies should prioritize diesel truck replacement, logistics electrification, emission-standard compliance, and modal shifts from road to railway.

Figure 3a,b presents the distribution of NOx and VOC emissions by vehicle category. Heavy- and medium-duty trucks and light-duty trucks dominated NOx emissions, contributing 48.1% and 27.9%, respectively, and collectively accounting for 76.0% of the total. Private cars were the leading source of VOC emissions (56.5%), followed by motorcycles (17.1%). H&M trucks were the main contributors to SO₂ (43.9%), while private cars dominated NH₃ emissions (87.3%). The contributions of various vehicle types to particulate matter were relatively balanced, with H&M trucks being the largest contributors at 26%. Private cars accounted for the highest share of CO emissions at 48.4%.

3.2. Energy Consumption Under the Different Scenarios

Based on scenario design, the LEAP model predicted the energy consumption distribution in 2030 and 2040 under two scenarios, forecasted the energy consumption trends of vehicles in Zhengzhou from 2022 to 2040, and revealed the energy consumption distribution of different vehicle types.

As shown in Figure 4a,b, under the existing policy scenario, gasoline and diesel consumption have always dominated, accounting for 67% and 20% in 2030, and 68% and 16% in 2040, respectively. The proportion of electricity increases slightly, from 6% in 2030 to 8% by 2040. As shown in Figure 4c,d, under the dual carbon scenario, the proportion of gasoline consumption decreases, from 62% in 2030 to 53% in 2040. The proportion of electricity consumption increases rapidly, from 12% to 23%.

As shown in Figure 5a, under the existing policy scenario, the total energy consumption for road transportation continues to rise from 2022 to 2040, reaching 19,641 ktce in 2040; under the dual carbon scenario, energy consumption first rises and then gradually declines, peaking at 13,699 ktce in 2030, with energy consumption in 2040 being 12,971 ktce, a 34% reduction compared to the existing policy scenario. The distribution of energy consumption by vehicle type across different scenarios in 2040 is shown in Figure 5b,c. Under different scenarios, the vehicle type with the highest energy consumption proportion is always private cars. In 2040, the energy consumption of private cars is 13,323 ktce and 9138 ktce under the existing policy and dual carbon scenarios, respectively, accounting for 68% and 70% of the total energy consumption.

3.3. Total Air Pollutant Emissions from Vehicles

Drawing on the 2022 road transport emission inventory for Zhengzhou and incorporating the scenario settings within the LEAP framework, this study projected the evolution of CO₂ and major air pollutants from 2022 to 2040, together with the changing contribution patterns of different vehicle categories. The simulated CO₂ trajectories for both scenarios are presented in Figure 6a, and these patterns closely correspond to the energy consumption trends illustrated earlier in Figure 6.

Under the existing policy scenario, CO₂ emissions continue to increase, reaching 45 Mt in 2030 and 55 Mt in 2040. Zhang et al. (2013) and Liu et al. (2018) and others’ studies on China’s transportation sector show that the anticipated growth rates are 6% and 4%, respectively [15,35]. The vehicle emissions in Zhengzhou have a similar growth trend, and the numerical differences may be caused by different research objects and statistical scopes. Under the dual carbon scenario, the carbon peak is achieved in 2030 with an emission of 36 Mt, and then it continues to decline, reaching 32 Mt in 2040. Compared with the existing policy scenario, CO₂ emissions under the dual carbon scenario are reduced by 20% in 2030 and 42% in 2040.

As shown in Figure 6b, under the existing policy and dual carbon scenarios, CO emissions decrease from 220 kt in 2022 to 217 kt and 111 kt in 2040, respectively. Compared to the existing policy scenario, CO emissions under the dual carbon scenario are reduced by 49%. The primary reason for this continuous decline is the progressive removal of vehicles that do not meet modern emission requirements and the growing share of vehicles certified under more stringent standards. This structural shift results in a steady decrease in CO emissions beginning in the base year. Hong et al. (2016) similarly concluded in their study that the sustained decline in CO and VOC emissions from 2015 to 2020 was driven by the large-scale retirement of yellow-label passenger vehicles and their replacement with higher-standard vehicles [36]. In the dual carbon scenario, CO emissions decrease rapidly due to the promotion of cleaner vehicles and the broader rollout of green transport initiatives.

As shown in Figure 6c, the emissions of VOCs remain largely unchanged from 2022 to 2040, under the existing policy scenario. In the dual carbon scenario, the emissions of VOCs in 2030 and 2040 are reduced to 24 kt and 18 kt, respectively. Compared with the existing policy scenario, the emissions of VOCs have decreased by 32% and 45%, respectively. The continuous reduction in VOC emissions is attributed to the continuous elimination of old vehicles and the replacement with vehicles that meet higher emission standards. Therefore, VOC and CO emissions exhibit similar trends under the two scenarios, consistent with the findings of Pang et al. (2022) [37].

As shown in Figure 6d, under the existing policy scenario, the NOx emissions from vehicles in Zhengzhou continue to increase slowly, reaching 73 kt in 2030 and 78 kt in 2040. Under the dual carbon scenario, NOx emissions continue to decline, with peak emissions in 2030 and 2040 being 53 kt and 44 kt, respectively. Compared with existing policy scenario, NOx decreases by 27% and 44%, respectively. As shown in Figure 6h, the NH₃ emissions in 2030 and 2040 are 3804 t and 4681 t, respectively, under the existing policy scenario. Under the dual carbon scenario, the peak emissions in 2030 and 2040 are 2900 t and 2512 t, respectively, representing a decrease of 24% and 46% compared with the existing policy scenario. NH₃ emissions peak in 2029 and then decline slowly, which is due to the reduction in the activity levels of trucks and buses caused by the elimination of old vehicles, the promotion of rail transportation, and green public transportation in the emission reduction policies. Xu et al. (2021) also found that the reduction in the activity levels of large buses and trucks led to a decrease in NOx and NH₃ emissions [38].

As shown in Figure 6g, under the existing policy scenario, the SO₂ emissions in 2030 and 2040 are 2741 tons and 3063 t, respectively. Under the dual carbon scenario, the SO₂ emissions in the peak year of 2030 and 2040 decrease to 2152 t and 1808 t, respectively, representing a reduction of 21% and 41%. Under the existing policy scenario, the SO₂ emissions increase with the increase in traditional fuel vehicles. Under the dual carbon scenario, with the increase in the proportion of electric vehicles and the implementation of higher vehicle emission standards, the SO₂ emissions reach their peak in 2026 and then continue to decline.

As shown in Figure 6e,f, the emission trends of PM₁₀ and PM_2.5 are similar. Under the existing policy scenarios, the emissions of PM₁₀ and PM_2.5 continue to increase, with the emissions reaching 2680 and 2568 t, respectively, in 2030, and 3254 and 3024 t, respectively, in 2040. In the dual carbon scenario, PM₁₀ and PM_2.5 in the peak year of 2030 are 1424.0 t and 1400.1 t, respectively, and 1199 t and 1176 t, respectively, in 2040. These reductions are primarily driven by the accelerated removal of high-emitting vehicles and the growing share of vehicles compliant with stricter emission standards. The projected divergence between CO₂ and regulated air pollutants is consistent with the different mechanisms governing these emissions. Tighter exhaust standards and accelerated retirement directly reduce CO, VOCs, NOx, and exhaust PM emission factors, whereas CO₂ remains closely linked to total energy demand, travel activity, vehicle efficiency, and electricity carbon intensity. Consequently, a city may achieve substantial reductions in regulated pollutants, while transport-related CO₂ continues to grow. The Zhengzhou results further show that policy ranking depends on the emission objective. EV promotion and green public transport provide relatively large CO₂ benefits, whereas old vehicle retirement, motorcycle phase-out, light-truck electrification, and emission-standard improvement provide greater proportional benefits for local air pollutants.

3.4. Vehicle Category Emission Profiles Under the Two Scenarios

Based on the paths and measures designed for each scenario, the LEAP model was employed to estimate the contributions of various vehicle categories to CO₂ and key air pollutants in 2022–2040 under the EPS and DCS. The results presented in Figure 7, Figure 8, Figure 9 and Figure 10 depict how different vehicle types influence the overall emission profiles across the two scenarios. Under the existing policy scenario, CO₂ emissions continue to rise throughout the projection period and do not reach a peak by 2030. In contrast, the dual carbon scenario yields a markedly different trajectory, with transport-related CO₂ emissions achieving a peak in 2030 and subsequently declining at a gradual pace.

Figure 7 shows the trend and composition of CO₂ emissions. Under the existing policy scenario, CO₂ emissions from all passenger vehicle categories increase continuously throughout the projection period. Private cars remain the dominant source, with their share rising from 67% in 2022 to 73% in 2040. Buses and taxis see their shares increase from 3% to 5% and from 2% to 4%, respectively. In contrast, the shares of H&M trucks and light trucks decline from 16% and 9% in 2022 to 10% and 4% in 2040, respectively, reflecting the combined effects of stricter emission standards and partial electrification of the logistics sector. Under the dual carbon scenario, private cars peak in 2027 (share: 77%), H&M trucks peak in 2028 (share: 14%), and light trucks peak in 2033 (share: 7%). Pursuant to Zhengzhou’s municipal policy on motorcycle management, all motorcycles are assumed to exit the market by 2025, thereby contributing no further tailpipe emissions from that year onward. Bus emissions remain modest as all 6316 buses are converted to electric vehicles by the end of 2022 per the provincial electrification mandate.

Figure 8 illustrates the evolution and sectoral distribution of NOx emissions. Under the existing policy scenario, enhancements in vehicle emission standards combined with measures such as promoting rail transportation lead NOx emissions from heavy- and medium-duty trucks to reach a peak around 2027, followed by a steady decline thereafter. The NOx emissions of private cars, affected by the improvement in vehicle emission standards and electrification, peak in 2028 and then gradually decrease. Due to the influence of urban logistics electrification, the emissions and proportions of light trucks continue to decrease. The emissions of buses and other vehicles continue to increase, with the proportions rising from 6% and 10% in 2022 to 17% and 29% in 2040. Under the dual carbon scenario, the emissions of six vehicle types, including trucks and private cars, continue to decrease, except for those of other vehicles. Due to the strengthened policy of urban logistics electrification, light fuel trucks exited the market by the end of 2025 and did not generate local emissions. Due to the strengthened policy of green public transportation, all buses were replaced with electric vehicles by the end of 2022 and did not generate local emissions.

The emission trend and composition changes in VOCs and PM_2.5 are shown in Figures S1 and S2, respectively. Under the existing policy scenario, VOC emissions in Zhengzhou exhibit a slow declining trend, with private cars (59%) and motorcycles (20%) emerging as the dominant contributors by 2040, while the share of buses rises markedly from 4% in 2022 to 14%. Light-duty trucks show a continuous decline because of the widespread use of clean energy vehicles. In contrast, under the dual carbon scenario, total VOC emissions decline more substantially, with private cars (91%) and heavy-duty trucks (6%) remaining the primary sources by 2040, as motorcycles, buses, and light-duty trucks are fully electrified. The evolution of PM_2.5 emissions is closely tied to the combined dynamics of vehicle electrification and emission-standard upgrades. Under the existing policy scenario, PM_2.5 emissions from private vehicles peak in 2031 and subsequently decline as the fleet transitions to higher emission standards. Under the dual carbon scenario, the early exit of motorcycles, fuel buses, and light fuel trucks (all by 2026) substantially reduces PM_2.5 emissions, a finding consistent with Pang et al. (2022) [37] who reported similar patterns in Lanzhou [38]. The residual PM_2.5 in both scenarios is increasingly dominated by non-exhaust sources (tire wear, brake abrasion), which are not captured in this study and represent a growing area of concern as EV penetration increases—since EVs reduce exhaust PM_2.5 but do not eliminate non-exhaust PM emissions [39].

3.5. Analysis of the Peak Year (2030) in Zhengzhou

Under the dual carbon scenario, CO₂ emissions can peak by 2030. This section focuses on the peak year and analyzes and discusses the energy consumption and CO₂ and air pollutant emission characteristics of the transportation in Zhengzhou City. It quantitatively assesses the contribution of the control paths to the reduction in CO₂ emissions and evaluates the synergy of different emission reduction measures on the reduction in CO₂ and air pollutants through the elasticity coefficient.

As shown in Figure 9, the total energy consumption for road transportation in Zhengzhou in 2030 is 13,699 ktce. The largest proportion is gasoline, accounting for 60.8%, followed by diesel at 20.4%. The proportion of electricity increases the fastest, rising from 0.6% in 2022 to 18.8% in 2030. Private cars make the highest contribution to energy consumption, at 70.4%, followed by the freight sector at 21.1%. The total CO₂ emissions in 2030 are 36 Mt, with private cars being the main source of CO₂ emissions, contributing 78.4%. In 2030, the total emissions of CO, VOCs, NOx, PM₁₀, PM_2.5, SO₂ and NH₃ from vehicles in Zhengzhou are 138.84 kt, 23.93 kt, 52.77 kt, 1424.74 t, 1400.1 t, 2152.16 t and 2900.77 t, respectively.

Compared with the EPS, CO₂ emissions under the DCS are approximately 9 Mt lower in 2030, corresponding to a reduction of 20%. Figure 10a further decomposes the formation of the 2030 peak-year emissions by considering road transport growth and the mitigation contributions of different policy pathways. Among all mitigation measures, the policies promoting electric vehicle adoption and advancing green transportation systems provide the largest contributions to CO₂ abatement in the peak year, accounting for approximately 42% and 32% of the total reduction, respectively.

At present, air pollution prevention and control require coordinated reductions in PM_2.5 and O₃. This necessitates coordinated reductions in NOx and VOCs. As shown in Figure 10b,c, in the dual carbon scenario, 2030 can achieve a reductions of 20 kt and 11 kt for NOx and VOCs compared to the existing policy scenario. The implementation of multiple policies has made a significant contribution to the reduction in NOx. Policies such as the elimination of motorcycles, the elimination of old vehicles, the encouragement of electric vehicles, and the electrification of urban logistics have the most significant effects, with their contribution rates to the peak-year reduction being 27%, 20%, 14%, and 14%, respectively. For the VOCs, the electrification of urban logistics, emission-standards updates, and the conversion of public transportation to railways have the largest reduction contributions, with their contribution rates to the peak-year reduction being 39%, 25%, and 21%, respectively.

The ELS analysis reveals the degree of synergy between CO₂ and air pollutant reductions under each measure (

Table 1. Elastic coefficients (ELSs) for the carbon peak scenario.

Scenario and Policy	ELSCO2/VOCs	ELSCO2/NOx	ELSCO2/NH3	ELSCO2/PM2.5	ELSCO2/PM10	ELSCO2/CO	ELSCO2/SO2	ELSCO2/Pollutants
Carbon peak scenario	0.60	0.54	0.70	0.46	0.44	0.54	0.68	0.54
EVP	1.74	11.22	0.54	0.98	1.07	1.46	1.07	1.81
REG	NA	NA	NA	NA	NA	NA	NA	NA
OVE	0.21	0.20	3.31	0.08	0.07	0.20	1.59	0.20
ME	0.04	0.77	0.72	0.26	0.28	0.04	0	0.06
IES	0	0	0	0	0	0	0	0
ELT	0.23	0.09	0.65	0.45	0.49	0.17	0.18	0.15
TR	1.06	0.27	2.52	1.27	1.39	0.74	0.37	0.54
GPT	1.91	2.29	0.81	1.11	1.03	1.88	1.46	1.91

NA indicates that the elasticity coefficient is undefined because the denominator is zero. EVP = EV promotion, REG = Renewable energy generation, OVE = Old vehicle elimination, ME = Motorcycle elimination, IES = Improve emission standards, ELT = Electrification of light trucks, TR = Transit rail, GPT = Green public transportation.

). Two policies achieve only single-pollutant benefits: the clean energy grid connection policy reduces CO₂ exclusively (ELS = undefined, since Δφn = 0), while emission-standard upgrades reduce only criteria pollutants (ELS = 0). All remaining policies demonstrate synergistic co-reduction effects (ELS > 0). Under the overall dual carbon scenario, the average ELS across all pollutants is 0.54, ranging from 0.44 to 0.70. The highest synergy is observed for NH₃ (ELS_CO2/NH3 = 0.70) and SO₂ (ELS_CO2/SO2 = 0.68), suggesting that CO₂ reduction measures in the dual carbon scenario are particularly effective at simultaneously reducing these two pollutants.

At the individual policy level, the EV promotion policy yields the highest ELS (ELS_{CO2/pollutants} = 1.81), indicating a proportionally greater effect on CO₂ than on local air pollutants—consistent with the fact that EVs eliminate tailpipe pollutants entirely, while CO₂ reductions depend on the carbon intensity of the electricity grid. In contrast, motorcycle phase-out generates a very low ELS (0.06), reflecting motorcycles’ disproportionately high contribution to VOCs and CO relative to CO₂. Similarly, light truck electrification (ELS = 0.15) and old vehicle phase-out (ELS = 0.20) both preferentially reduce criteria pollutants over CO₂, because stricter emission standards directly constrain exhaust pollutants but have limited impact on fuel-linked CO₂ emissions. These results highlight that a portfolio of policies, combining high-ELS measures (EV promotion, green transport) for CO₂ leadership with low-ELS measures (phase-outs, standard upgrades) for air quality leadership, is necessary to maximize synergistic co-benefits.

The elasticity results should not be interpreted as indicating that a larger coefficient is always preferable. A high ELS indicates that the proportional CO₂ reduction is large relative to the pollutant reduction, whereas a low positive ELS may identify a measure that is particularly effective for air pollutant control. Measures with high and low ELS values are therefore complementary. Coordinated policy design should combine carbon-oriented measures, including vehicle electrification, public transport, freight modal shift, and electricity decarbonization, with pollution-oriented measures, including accelerated retirement of high-emitting vehicles and tighter emission-standard enforcement.

3.6. Analysis of the Relationship Between the Penetration Rate of EVs and Charging Infrastructure

By summarizing the historical data of vehicle population, EV penetration rate, and charging infrastructure in Zhengzhou from 2019 to 2024, combined with existing policy scenarios and dual carbon scenarios, the future demand for charging piles in Zhengzhou is predicted based on EV population data for 2030 and 2040.

Table 2. Data on charging piles in Zhengzhou from 2019 to 2024 (unit: thousand).

Years	2019	2020	2021	2022	2023	2024
Vp	4400	4300	4870	5220	5650	5930
EVp	110	120	140	260	380	520
VCs	12.6	20	23	44	69	214
VCR	8.6	6.1	6.1	5.8	5.5	2.4

Vp = Vehicle population, EVp = EV population, VCs = Vehicles to charge, VCR = Vehicle-to-charge ratio.

presents historical data on vehicle population, EV penetration, and charging infrastructure in Zhengzhou from 2019 to 2024. The total vehicle population grew from 4.4 million in 2019 to 5.93 million in 2024, at an average annual growth rate of 5.1%. Over the same period, the EV stock expanded dramatically from 108,000 to 520,000, raising the penetration rate from 2.5% to 9.2%. The number of charging piles increased from 12,600 in 2019 to 214,000 in 2024, while the vehicle-to-charger ratio decreased from 8.6 to 2.4. This indicates a substantial improvement in charging infrastructure availability, although the 2024 ratio remained slightly above the target value of 2:1. Under the existing policy scenario, the demand for charging piles will reach approximately 370,000 by 2030 and 1.06 million by 2040. Under the dual carbon scenario, the demand rises to approximately 610,000 by 2030 and 1.73 million by 2040 (

Table 3. Data on charging piles in Zhengzhou in 2030 and 2040 under the predicted scenario (unit: thousand).

Years	EP2030	EP2040	DC2030	DC2040
Vp	7570	7630	7520	7290
EVp	740	2110	1220	3450
VCD	370	1055	610	1725
VCRt	2	2	2	2

VCD = Vehicle-to-charge demand, VCR_t = Target vehicle-to-charge ratio.

).

By accelerating the construction of a well-distributed charging network with a car-to-pile ratio, a public charging intelligent management and control service platform that covers the entire city, features comprehensive functions, and enables “car–pile–network” interconnection will be formed to enhance the charging efficiency of new energy vehicles. Research indicates that the intelligent management and control service platform can not only reduce user charging waiting time by more than 30% but can also increase the utilization rate of power distribution equipment by 22–38%.

The estimated charging demand represents an aggregate planning requirement rather than a detailed engineering deployment plan. Increasing the number of charging piles from approximately 214,000 in 2024 to 610,000 in 2030 would require sustained infrastructure expansion and coordination with the local distribution grid. Its implementation may also be influenced by the charger type, utilization, land availability, and spatial demand distribution. Therefore, delays in charging infrastructure or grid upgrading could affect the pace of EV adoption. The DCS may also involve additional investment in vehicle replacement, charging facilities, grid reinforcement, and operation and maintenance. These economic factors are not explicitly included in the present emissions-focused scenario analysis and should be considered in future assessments of policy feasibility and cost-effectiveness.

4. Conclusions

This study developed an integrated city-level framework linking a co-source inventory of CO₂ and seven air pollutants, LEAP-based scenario modeling, policy contribution analysis, elasticity-based co-benefit assessment, and charging infrastructure demand estimation. Unlike assessments that examine carbon mitigation, air pollutant control, and electric-mobility infrastructure separately, the framework evaluates these dimensions using a consistent vehicle classification and scenario structure.

In the 2022 base year, Zhengzhou’s road transport sector consumed 10,178 ktce of energy and emitted 27.8 Mt of CO₂. Private cars were the dominant source of CO₂, contributing 66.7%, whereas heavy- and medium-duty trucks and light-duty trucks collectively accounted for 76.0% of NOx emissions. These contrasting source profiles indicate that passenger and freight transport require differentiated control strategies.

Continuation of existing policies is insufficient to achieve a road transport CO₂ peak before 2030. Under the EPS, CO₂ emissions increase to 45 Mt in 2030 and 55 Mt in 2040. Under the DCS, emissions peak at approximately 36 Mt in 2030 and decline to 32 Mt by 2040, corresponding to reductions of 20% and 42% relative to the EPS. EV promotion and green transport development provide the largest contributions to peak-year CO₂ mitigation, accounting for 42% and 32% of the total reduction, respectively.

The policy-specific analysis shows that no single measure simultaneously maximizes CO₂ and air pollutant reductions. The elasticity results support a differentiated and staged policy portfolio. In the short term, accelerated retirement of old high-emitting vehicles, motorcycle phase-out, tighter emission-standard enforcement, and light-truck electrification should be prioritized to achieve rapid reductions in NOx, VOCs, CO, and exhaust PM. In the medium and long term, large-scale private-vehicle electrification, green public transport, freight modal shifts, and electricity-sector decarbonization should become the principal measures for constraining CO₂ growth and maintaining the post-peak decline. Measures with high and low ELS values should be regarded as complementary components of an integrated portfolio rather than competing alternatives.

Charging infrastructure and power-grid capacity are enabling conditions for the mitigation pathway. Under the DCS, approximately 1.22 million EVs and 610,000 charging piles would be required by 2030 to achieve a vehicle-to-charger ratio of 2:1. Infrastructure planning should consider not only the number of chargers but also their spatial allocation, utilization, fast- and slow-charging composition, grid-connection capacity, land requirements, and investment needs. Delayed deployment could constrain EV adoption and weaken the projected emission benefits.

This study is subject to several limitations. The air pollutant inventory covers exhaust emissions and excludes upstream and non-exhaust emissions. Non-exhaust particulate emissions, including tire wear, brake wear, road-surface abrasion, and road-dust resuspension, are not included. The scenario results depend on assumptions concerning transport activity, electrification, electricity carbon intensity, technology adoption, and policy implementation and do not explicitly represent abrupt economic or supply-chain shocks. Future research should incorporate non-exhaust particulate matter, upstream air pollutants, spatially resolved charging loads, grid constraints, investment costs, and probabilistic uncertainty. Subject to these limitations, the Zhengzhou case demonstrates that coordinated vehicle, transport structure, emission control, power sector, and infrastructure policies are required to achieve urban transport carbon peaking and air pollution co-control.